NetCDF Territorial Classification and Python Colab Scripts Based on Welch Frequency-Spectrum Analysis of SbAI Signals Between 0 and 0.5 Cycles per Year in Southern South America (Argentina, Chile, and Uruguay). (FS-WEL-2-inf)
Otros responsables
Fecha
Publicador:
Consejo Nacional de Investigaciones Científicas y Técnicas
Fecha de depósito:
27/05/2026
Fecha de creación:
12/5/2026
Clasificación temática
Palabras clave
Resumen
This notebook implements a full pipeline for spectral analysis and clustering of FIR-filtered SbAI data:
- Computes Welch power spectra per pixel (restricted to periods > 2 months and cutoff at 2 years).
- Applies KMeans++ clustering (RobustScaler preprocessing, n_init=10).
- Generates classification maps with geographic boundaries from a geopackage.
- Produces cluster centroids and representative spectra.
- Provides elbow method plots to evaluate the optimal number of clusters using R², relative reconstruction error (%), and inertia (SSE).
Información Técnica
The workflow consists of:
1. Mounting Google Drive and setting the working directory (`base_dir`).
2. Loading FIR-filtered dataset (`netcdf_sbai_025_lowpass_fir.nc`).
3. Computing Welch spectra per pixel:
- Frequency resolution increased with `nfft`.
- Restriction to periods > 2 months and ≤ 2 years.
- Saving results as NetCDF (`welch_fir_lowpass_2y.nc`).
4. Applying KMeans++ clustering (5 clusters, RobustScaler, n_init=10):
- Stacking spectra into pixel × frequency matrix.
- Computing metrics: explained variance (R²), relative reconstruction error (%), inertia (SSE).
- Saving classification results (`classification_kmeans_lowpass2y.nc`).
5. Plotting:
- Geographic cluster map with `canvas.gpkg`.
- Centroid spectra (smoothed).
- Representative spectra closest to centroids.
6. Elbow method test:
- Iterates over k=2…10 clusters.
- Computes R², relative reconstruction error (%), and inertia (SSE).
- Generates separate plots for each metric.
Colecciones
Condiciones de uso
Las buenas prácticas científicas esperan que se otorgue el crédito adecuado mediante una citación. Utilice un formato de citación y aplique estas normas de reutilización:
Licencia Creative Commons
Excepto donde se indique lo contrario, la licencia de este ítem se describe como by-nc-sa
Archivos
ARCHIVO
NOTAS DE USO
TAMAÑO